This application proposes a structural ecosystemic intervention for HIV medical adherence (SETA) in recently sober women. Recently sober is defined in this proposal as DSM IV drug dependence or abuse within the last year and but not in the last 60 days. The 4-month intervention targets women, their families, and their social networks as the building blocks for the infrastructure to support HIV medical adherence, reduction in HIV transmission risk behaviors and drug abuse relapse prevention. An important part of this infrastructure is a constructive relationship between the HIV infected women and their health care system. This is buttressed with attention to 1) reducing the negativity former drug users experience from their families, 2) establishing firm boundaries to separate the women from drug using family and friends who can frequently trigger relapse to drug use, 3) for women with children, restoring the women=s role as a parent, improving her parenting abilities, and enhancing the support she receives for parenting functions and 4) developing well integrated, multiply connected pro-social support systems to reinforce the recovery process and maintain healthy behaviors including HIV transmission risk and HIV medical adherence. The study will enroll 196 women and randomly assign them to either the SETA intervention, or an HIV health group designed to match SETA for attention. Women are assessed at 2-month intervals for a period of 12 months. HIV medical adherence is measured by self-report, MEMS CAP, and viral load. The SETA intervention is hypothesized to improve HIV medical adherence relative to the HIV health control group. The hypothesized mediating factors are: family functioning, social support, patient-doctor interactions, health beliefs, and drug abuse relapse. The effects of the intervention and the hypothesized mediators will be tested using Hierarchical Linear Modeling. This will be the first award for this investigator.